On Sep 11, 2005, at 3:51 PM, J. C. O'Connell wrote:
My sketch was a simple sketch ( that obviously not
a real lens design with two convex lenses) and I
explaied I was showing the ACTIVE area of the rear
element so it does not matter where the nodal point
because if the entire active area gets closer to the
sensor then the angles to the corners of the sensors
get further away from perpendicular/ideal.
This makes no sense.
In a lens with a set of rear elements designed to correct light path
to orthogonal, a large percentage of the rear element of the lens is
*always* active. That is the point of the design: direct the light to
be orthogonal to the sensor. If the rear elements are close to the
imaging plane, and far from the nodal point, the correction is small
and most of the rear element is being used. If the rear elements are
far from the imaging plane and close to the nodal point, a smaller
percentage of the rear elements are being used and the correction
possible is reduced.
... a condenser enlarger head does: it
positions a collimating lens group very close to the film plane in
order to make the light pass evenly through all points of the
negative, right to the corners, and oriented orthogonally through the
film so that a flat field imaging objective (the enlarging lens) will
exhibit very little light falloff at corners and edges.
I totally disagree with the englarger light house
because the output of an enlarger condensor assembly
is PARALLEL light rays going to the film form a point
light source. A camera, digital or otherwise has a
POINT SOURCE image formed at the film/sensor from a point
source REAL OBJECT, in other words the output of a camera
lens is an image of the real object formed on the film/sensor
while the output of an enlager condensor lamphouse is completely
different, its NOT forming an image of the enlarger lamp,
its forming a cylinder of parallel light rays instead of
an image at the film.
The use of a condenser enlarger as example is illustrate simulating a
point light source at infinity such that the ray trace over the area
of the film would be parallel. This is indeed the way light coming
from a point source at infinity would be oriented. In the camera lens/
sensor system, the point source can be seen as the lens' nodal point.
A large diameter element at the rear of a lens designed for the
digital sensor helps in promoting this even illumination of the
entire sensor area. Placing this rear lens group close to the sensor,
relatively distant from the nodal point, allows the strength of the
elements to be lower and thus promotes less distortion from the
correction.
You are overlooking that the "diameter" of the rear element
is not "fixed" and it gets smaller in its active area ( optical
path), quite small in fact at small fstops like f11/16 so
that is changing with lens settings and cannot be maintained
constant...So if the advantage of the large rear element
is there its not constant and the angle at which the light
rays hit the sensor corners is worse when the lens is stopped
down.
See above. Perhaps I'll draw a diagram or two for you.
Secondly, I totally agree that increasing
the nodal point away from the sensor while
maintaining the same focal length will help
the digital sensor / lens interface with respect
to keep the lens rays more parallel incidence
at sensor plane but there is a heavy price
for that , the retrofocus lenses that do that
are far larger, heaver, worse optically, and
more expensive than if you don't need to do that.
That's why the Pentax and other cameras that
use APS sensors in old FF 35mm body designed
lenses are at a disadvantage, the 45.5mm sensor
plane to lens flange distance is way too large
relative to the small format (APS),
I'm not exactly sure what you're trying to say here.
Yes, inverted telephoto designs are typically more complex, heavier
and more expensive than non-inverted-telephoto designs. They are a
result of the need for more clearance with SLR bodies as focal length
is reduced. On the other hand, evenness of illumination is typically
better with inverse telephoto designs.
As far as I'm aware, nearly all modern 35mm and shorter focal length
lenses designed for 35mm and digital SLRs are inverse telephoto
designs. Most of the better, modern wide angles used for rangefinder
cameras are as well, because the even illumination is useful. Only a
few are not, and those generally demonstrate corner/edge falloff to a
greater degree.
The register distance could be shorter in dedicated lenses for a DSLR
due to the shorter mirror required, but the whole point of using the
current register distance is to enable use of existing lens and mount
designs. IF, however, you're designing a mount and lens from scratch
for a digital sensor, you'd use a wider diameter mount with a shorter
register. This lets you place large diameter, corrective rear
elements closer to the sensor without having to go as far with moving
the nodal point forward through inverted telephoto design.
Godfrey
